Mr. Using Chairman the assum , I ption am plth eas ated wag to essubmit or othe tr his bes nte afit tement s do no for t t ad he jus r0cord t, _o that on th0 any 0s cot_t imincro at0sa_ ofo_ th0 or Table 1 Table 2 Table 4 savings due to the mandate are borne by the employer, we find that the Clinton T-91 T impact he follo of wit ng he Clin tables ton comp Adminis are trth ation e job 's loss health or gain careestimates reform proposal resulting on frojob m san . The employer Employ mandate ee Number of Workers per Unit Adjustment Factor: Adn using Estimates Estimates _inistration's a variet of of y Job Job ofprLosses ass Losses oposal umpti(- ( o- would ns. ) ) o orr Gains Gains The create due du tables ebetween to to an dem an Employe Empl ons 631,000 tro ate yer r and M the Man and 661,000 sensitivity da ate te Using Using net new oVarious V f ario these jobs us estimates Assu A (see ssump m tables pti tio on ntss o2, 3, Benefit Research Institute (EBRI), a nonprofit, nonpartisan public policy research organization, Determination of Employer per Worker Contribution, by State, 1991 assumpti is and dedicated 4), depending ons o on n Costs Costs to used. providing to toon Ea Employe Employe th che cpremiums. ol ob umn r rje s, s,ctive Caps Caps repanal resents on App onysEmploye Employe irs oximately the of health rr resuCont Contribu lts600,000 care o ribu f simulati ti ti and ons ons, jobs , other ons would an and d using wo Changes Changes r be k lost force a particula fin ir nom iWages Wages ssue r those s. ,,set Toofthat EBRI Tabulations of the March 1992 Current Population Survey (CPS) assumptions: By Employer Size end employe , thirss testimon who face y refl new ectscosts, the and reseaapp rchBy roximately EBR Family I has Income compiled 1.3 million on new the jobs posswould ible effec be ts created of an by Couples with Couples with those employers who experience lower costs as a result of the mandate. employer mandate under the Administration's health care reform proposal. Using a variety of no children children (1) (2) (3) (4) assumptions Column (1): , These EBRI estimates has produced assume a range a family of estima premium tes onofthe $5,9 emplo 00 an yd ment an indivi effects dual of premium the of (1) (2) (3) (4) The $2,40assump 0. Alltifo amilies n thatco wages mpoH seigh dor o of the m rore benefits than owould ne H pe igh rso not n change pay the after fami Cllin yato mandate p nremium. is It enacted is assumed , or Administration's employer mandate of between 661,000 net jobs created to 158,000 net jobs lost. National Total 1.43 1.53 High High Clinton Premiums Premiums Premiums Clinton announced, is highly unrealistic. It represents an upper bound on change in the employment that employers contribute 90 percent of these premiums. It is assumed that wages and other However, the Administration's proposal for determining the employer contribution is likely to Total Premiums Premiums Premiums Clinton Numbe State r of Current Clinton Clinton Premiums that compensati would on result do n fr oom t change an employer with themandate. implementati Aton the oo th fethe r extreme employeris th man e assumption date. that reduce the cash income of lower wage workers and increase the cash income of higher wage Family Current Clinton Clinton Premiums workers. wages adjust, upward or downward, to completely absorb the costs of an employer mandate. In Employees Methodolo_¢ Methodolob'y Methodolobn/" Wa_es Adjust Alabama 1.31 1.53 Income Methodolo_ Methodolob']¢ Methodolob_ Wages Adjust U this ndecase, r 10 the employers' -6 costs 77,938 are unchanged and13,253 their demand for 1labo 6,97r 5 would be unch -36,618 anged, Column (2): These estimates assume a family premium of $5,900 and an individual premium of Alaska 1.45 1.40 Under $5,000 -592 -43 -43 -22 except for those workers at or near the minimum wage. Those workers' wages could fall to the $2 ES ,4T 00. IMATE All fami S liOF es cJ oO mB posed LOSS of m D oUE re than TOoAN ne pers EM onPLOYER pay the fami MA lyND premium. ATE Employers Arizona 1.34 1.45 10 to 24 -431,349 53,068 55,793 -24,485 minimum contribute wage an amo but untnoeq fu ur al thetro , 8 so 0 pe employe rcent rof s the of minimum premium wage dividedworb ke y ran s would adjustment face hi fact ghe or r. costs This Arkansas 1.33 1.62 $5,000-10,000 -195,773 4,568 4,654 -11,649 California 1.37 1.44 25-99 as a result of the mandate.-507 Using ,764 the assumption153that ,547 wages adjust 157completely ,634 except -30,125 for those adjustment factor is equal to the number of workers in families with two adults divided by the Analysts must make three basic assumptions Statement in order to arrive at an estimate of the impact of workers $10,000-15,000 near the minimum -383,wage, 758 we estimate 49,820 that an employer mandate 50,149 would cause -21,77 about 6 number of families with two adults. Employer contributions are capped at 3.5 percent of payroll an employer mandate on jobs. First, they must determine the cost of the benefit being mandated. 10(0499 -397,714 89,662 94,954 -22,088 Colorado 1.41 1.48 158,000 jobs to be lost. This estimate would be larger if higher premiums were used in the for individuals t employed by employers with fewer than 25 employees and having wages of Second, they must use an estimate of the sensitivity of the demand for labor to a change in the Connecticut 1.65 1.59 $15,000-20,000 -323,126 94,219 94,835 -16,352 simulation. 500-999 -145,630 52,568 54,510 -6,841 less than $24,000, increasing to 7.9 percent for employers with more than 75 employees. It is cost of labor. Finally, and most importantly, they need to make an assumption about the Delaware 1.48 1.59 $20,000-30,000 -543,634 169,641 172,564 -34,366 assumed District thatofwages Columbia and other compensati1.49 on do not change with the implementati 1.53 on of the operation of the labor market and the speed with which it adjusts. Before the 1,000+ -821,772 269,092 281,646 -48,109 Florida 1.32 1.55 emp Using loyeth r e mandate. national average premiums estimated by the Clinton Administration, a reasonable $30,000-50,000 -790,316 262,313 268,435 -42,448 range Total for the impact of-2 th ,982,166 e employer mandate on 63jobs 1,190is between 661,000 661,512 net jobs created -168,265 and In the long run, mandating that a benefit be offered does not affect the labor market. In general, Georgia Subcommittee 1.39 on Health 1.60 $50 158000 ,000-100,000 jobs lost. Higher -656 pre,mi 840 ums would reduce 73,821 the number of jobs 89,38 cr7 eated at one extreme -36,460 and Column (3): These estimates assume premiums to be $1,800 per single adult, $3,630 for a single So it does urce: nEB ot RI affect simul tha etion ins trinsic using value the of Maa rcwo h rke 1993 r S to upp anleemploye ment rto th ore ch Can urr ge ent thP eop nu umbe lation r of Su wo rvey rkers Hawaii 1.68 1.62 in adu crlease t with thechi numbe ldren,r of $3,jobs 600 flost or aat coth up eleoth w eirth extreme, no childrbut en, itan is d unlikely $4,200 fo th rattwth oe adults range w would ith Committee on Ways and Means (CPS). available. Idaho Employer It does Contributions affect the composi are tid oe nriv 1.29 ed of total from compensa the National tion, Medical which may Expenditur 1.52havee impo Survey rtant, $100,000 or more -88,127 -23,148 -18,468 -5,192 exceed Illinois one-half of I percent of the work 1.55 force in either direction. 1.54 chil adjust dre en. d for All in families flation, co and mpoimputed sed of mto orethethan CPS. one person pay the family premium. Employers policy and economic implications, but it will not affect the number of jobs. U.S. House of Representatives Indiana 1.42 1.58 Total -2,982,166 631,190 661,512 -168,265 contribute an amount equal to 80 percent of the premium divided an adjustment factor. This It is important to note, however, that an employer mandate results in a redistribution of income. adjustment factor is equal to the number of workers in families with two adults divided by the Sou In the rce:shEBRI ort and simula medium tions r using un, ho the weve Ma r, rch ma , rk 1993 ets Supplement may not adjutst o th comp e Cu let rrel ent y. Population In these cases Sur ,vey the Iowa 1.49 1.68 Higher income workers are more likely to have coverage currently than lower income workers. number of families with two adults. Employer contributions are capped at 3.5 percent of ( cos CP tS). of th Employe e mandat r e cowill ntribu bti eodivided ns are a de mong: rived wo from rkerth s einNa th tieonform al Medical of lower Expe wages nditu and re /orSurv fe ey wer , Kansas 1.50 1.58 Hearing on Many of these workers will find that their wages or other benefits increase as a result of the adjusted for inflation, and imputed to the CPS. job pay sroll ; emplo foryeir ndiv s iin dua th le s form t empl of olow yeder by profits; emploand yers con w sium th er fewe s r in than the form 25 empl of high oyees er p and riceshaving . Kentucky 1.28 1.34 mandate. Louisia Conve na rsely, those workers who 1.29 do not now receive health insurance 1.50 through their Simil wages arly of , less the than benefi$24 ts ,000 of ,a inc cap reasing on emplo toy7 er .9 percent contribution for semplo for yers healtwith h benefi morte s than will 7 be 5 empl divided oyees. Maine 1.41 1.54 employer tend The to be Impact lower-income on Jobswo of rkethe rs. Administration's These lower income Health workers Security will find th Act at their It among is assuwo med rkersthat , emplo wages yers,andand other cons cum ompensati ers. o Th ne ad ctual o notsha change re of th with e burd the en imp orlementati benefit on is likel of y the to Table 3 real wages will fall or their jobs will become less secure, or both, as a result of the mandate. employer mandate. vary by geographic region, occupation, industry, and size of employer. Maryland 1.63 1.61 Many of these workers will find they now have a health insurance benefit that they did not Estimates of Job Losses (-) or Gains due to an Employer Mandate Using Various Assumptions Mass ac husetts 1.56 by 1.49 have before, but 30 million of those who do not now get coverage from their employer have on Costs to Employers, Caps on Employer Contributions, and Changes in Wages, Column (4): All assumptions are identical to Column (3) except that wages are assumed to adjust As a nonpartisan, nonprofit research institute, EBRI's role in the health care debate is to Michigan 1.43 1.49 coverage Minnesota from a spouse's planBy orAve presently rage 1.54 Hou purrscha Wo se rked cover Page er Was eekindividuals. 1.68 While the benefit to completely eliminate the change in total compensation, except for workers close to the provide objective information on the tradeoffs inherent in all the health care reform proposals. William S. Custer, Ph.D. Mississippi 1.23 1.56 package under the Administration's proposal may be more generous than the plans these minimum wage. For these workers, wages are assumed to fall to the minimum wage, and To that end, we have developed our own micro-simulation model to: understand how Director of Research individuals now have, it is unclear if they will feel they are better off as a result of this (1) (2) (3) (4) employers will bear any additional costs due to the mandate. researchers arrive at widely different estimates, examine the sensitivity of these estimates to Missouri 1.48 1.67 reform. Employee Benefit Research Institute High High Clinton different assumptions, and understand the weaknesses in the methodologies employed. By Montana 1.32 1.55 Premiums Premiums Premiums Clinton examining Nebraska the assumptions and methodologie 1.45s used in the various estimates 1.73 of the impact of an Average Nevada Hours Current 1.48 Clinton Clinton 1.52 Premiums employer mandate, we can place reasonable bounds around the likely impact of an employer New Hampshire 1.61 1.54 mandate Per Weekon employmenMe t. thodolobry Methodolob_¢ Methodolob_a¢ Wa_es Adjust Washington, D.C. 10-20 hrs -68,160 18,052 21,179 -29,495 New Jersey 1.67 1.49 The Employment Policies Institute (EPI) released a report that contained estimates of 3.1 New Mexico 1.25 1.48 20-30 hrs -913,249 61,478 64,236 -51,974 million New jobsYork lost as a result of an employer1.46 mandate. The study was done befo 1.45 re the Clinton 4 November 1993 30-35 North hrs Carolina -404,856 1.43 35,094 37,098 1.60 -26,474 Administration released its health plan, so it did not incorporate the actual mandate North Dakota 1.33 1.68 contained in the Administration's proposal. EPI assumed that benefits would cost $2,400 for an Full Time -1,595,900 516,566 538,999 -60,322 individual and $5,900 for a family. They further assumed that employers would pay 90 percent Ohio 1.40 1.52 Total -2,982,166 631,190 661,512 -168,265 of the premiums. Finally, they assumed that wages would not change as a result of an employer Oklahoma 1.41 1.43 mandate, so that employers would bear the full costs. They used the March Supplement to the Source:Oregon EBRI simulations using the March, 1.34 1993 Supplement to the Current 1P .53 opulation Survey 1991 Cu Pen rrent nsylvan Popula ia tion Survey to derive 1.48 their estimate. Using similar, bu 1.47 t not identical, (CPS). Employer contributions are derived from the National Medical Expenditure Survey, Rhode Island 1.57 1.47 methodology and EPI assumptions on the premiums, we arrived at an estimate of 2.9 million jobs adjusted for inflation, and imputed to the CPS. lost due to a mandate, using the March Supplement to the 1993 Current Population Survey. South Carolina 1.46 1.63 South Dakota 1.36 1.73 In contr Tennessee ast to the assumptions used in the1.40 EPI and most other studies of the1.59 effects of an Texas 1.36 1.54 employer mandate, the Clinton Administration's proposal creates four premium categories: Utah 1.49 1.56 single adults, single adult families, couples with no children, and couples with children. The employer's contribution for each worker in each category is 80 percent of the premium divided Vermont 1.49 1.65 by an adjustment factor. The adjustment factor is the number of workers in each category Virginia 1.58 1.65 divided by the number of families of that type in each category in the alliance. Thus, the Washington 1.28 1.51 West Virginia 1.16 1.35 actual costs to the employer for family coverage are lower than the costs to employers who Wisconsin 1.53 1.74 provide such coverage now. In fact, this methodology results in a subsidy for one-worker couples Wyoming 1.25 1.57 paid by two-worker couples. That is, the sum of the contributions (employer plus family) of two-worker couples will exceed the premiums for their premium category, while the sum of the contributions of one-worker families will be less than the premiums for their premium category. It is important to note that, as shown in table 1, the adjustment factor, and therefore the employer contribution, varies dramatically from state to state. The Clinton Administration's proposal also caps the costs employers must pay as a percentage The views expressed in this statement are solely those of the author and should not be attributed to the of payroll. At most, employers are required to pay 7.9 percent of payroll, although the amount Employee Benefit Research Institute, its officers, trustees, sponsors, or other staff. The Employee Benefit t No Research te the Institute differencis e a nonprofit, between this nonpartisan, estimation public assu policy mption research and organization. the Clinton proposal, which uses may be as low as 3.5 percent for firms with fewer than 25 workers and average wages of less average payroll of the employer, rather than the individual's wages and has a sliding scale than $12,000. Suite 600 for the premium cap that is 3.5 percent for employers with average wages of less than $12,000 2121 K Street. NW up to 7.9 percent for small employers with salaries of over $24,000. Washington, DC 20037-1896 202-659-0670 Fax 202-775-6312